E6600 temps on p5wdh

[Geekwannab]

Originally posted by: renozi
Geekwannab enable Intel Speed Step and set the power option in windows to Portable/Laptop. E6700 @stock, idle 31C, load 45C in everest. And this is on a hot day!
My ram does 4-4-4-8 right now @800 with 2.0v, but could do it with auto volts too...I just wanted to be safe and run 2.0v like spec'd.

The whole point of doing all this(stock everything first up) was to get a guage of how much OCing room do I have. The temps @ stock setting is always an indicator how far can you OC with that particlular processor as each processor varies with their OCing capabilites.So enabling speed step is not an option coz if you enable speed step you cant increase the Vcore. Thanks for your input anyway.

 

[Geekwannab]

Originally posted by: renozi
Geekwannab enable Intel Speed Step and set the power option in windows to Portable/Laptop. E6700 @stock, idle 31C, load 45C in everest. And this is on a hot day!
My ram does 4-4-4-8 right now @800 with 2.0v, but could do it with auto volts too...I just wanted to be safe and run 2.0v like spec'd.

I saw yoursignature and realized that there is one more important paramter to the whole equation, which is the BIOS version. You were saying that you were able to run your RAM at the rated timing,speed and voltage with the 1305 BIOS. Guess what, I took that as a cue and flashed my bios to 1305(was running 1101) and disabled the SPD timings and running my RAM at the rated speed,timing and voltage(DDR2 800 @4-4-4-12 2.1V) and VOILA SP2004 is running fine now(approaching 30 mins as I speak). So I guess with the 1305 BIOS maybe they have fixed the divider issue on the P5WDH. If thats the case this is good news to a lot of people out there who have been complaining about the poor divider "sensitivity" of the P5WDH and tweak their system a bit more to get more performance. This opens up a lot of OCIng possiblities.

 

[stevty2889]

Finaly got my motherboard yesterday. So far running at 43c max load using scythe ninja, at 2.84ghz on stock voltage. 8 hours SP2004 stable and still going. My board shipped with 801 bios, but it booted right up anyway, using 1201 bios now.

 

[Geekwannab]

SP2004 failed after running for about an hour and half. Anyway I am happy with my idle temps now

Idle Temps
Asus probe - 34C
speedfan - 35C
Everest -34C
Core temp-50C

The load temps .
Asus probe - 48C
speedfan - 48C
Everest -49C
Core temp-63C

Will drop the RAM speed to DRR2-533(1:1) with tighter timings(3-3-3-5) and lower voltage(1.9V) and will see what happens

 

[DJ Mike]

I'm going to buck the trend of high temperature readings myself now. My setup is a P5W DH mobo with E6600 at standard 266FSB.

I've fitted my Scythe Ninja three times. The first two occasions, I was seeing idle temps (in Core Temp) of between 52-56C per core, with load temps going way up to over 70C when running Orthos in Blend mode. On a cold startup, even the mobo's BIOS would report 31-32C for the CPU.

So third time's the charm, eh? I refitted the Ninja, only this time I rubbed a tiny bit of Arctic Silver 5 onto the HSF ban to prime it, then put the tiniest blob onto the middle of the CPU - didn't spread it out like I did the previous two times, just planted my Ninja straight on top to squash it out. Was a bit skeptical about doing it this way, but the results speak for themselves:

Cold start BIOS CPU reading was 21.5C (I thought I'd buggered my mobo up at first!). Idle "Core Temp" readings are between 36-40C , while the highest temp I recorded under load (using Orthos Blend mode again) was 51C - for the most part, temps stayed at 100% load stayed around 46-48C.

So there you have it - you can get the temps of your Conroe down, you just have to be willing to persevere (and in my case, take a chance on using the tiny blob of AS5 in the middle of the CPU!) ;-)

 

[Geekwannab]

It was real hard and too tempting not to OC my E6600 with stock HSF. So without any VCore increase(AUTO), I OCed my E6000 to 2.7G(300FSB) with my RAM doing DDR2-600
4-4-4-12 @2V and my temps remained almost the same.Idle around 35-36C in all of Asus probe/speedfan/everest and coretemp reading 49C.The load temps were 48-49C in all of Asus probe/speedfan/everest and core temp reads 65-67C(after almost 2 hours of SP2004). Now that I got my ninja I would mount it with AS5 and post the results.

 

[DJ Mike]

Your temperatures are.. concerning, but not outrageous. I was getting 71-72C at 100% load with Orthos taking readings from Core Temp (the only reading I actually care about - maybe the CPU reading is too high but I'd rather be safe than sorry). That was due to a poorly-installed Ninja.

Since you're using the stock HSF I expect that the 65-67C you're seeing is quite reasonable given the expectation Make sure you clean up the CPU thoroughly with isopropyl, Q-Tips and finally a lint-free cloth (to get the CPU squeaky clean and also to wipe away any lint from the Q-Tips) when installing your new Ninja! Also, I highly recommend priming the base of the Ninja with a dash of AS5 (then wipe away the excess with lint-free cloth). Then when you put the AS5 on the CPU, just use a tiny blob, barely the size of the small grain of rice - and DON'T spread it around - just plant the Ninja straight down on top, which will squish it outwards, and clip it to the holders. Worked wonders for me!

 

[spectre8]

I don't mean to thread jack at all, but I'm having what seems to be overheating issues with my setup.

E6400 (running at the speed its supposed to be, no adjustments needed)
2 x 512mb of Corsair XMS PC2-6400 (I adjusted the voltage to what it's supposed to be, 2.1V)
Asus P5WDH
ATI X850XT
550W Antec Truepower 2.0

I first realized there was a problem when I tried to benchmark this new system with 3dMark06 (demo) and not even 30 seconds into the first test, the system speaker beeped once.. then again about 15 seconds later, then again, and finally the test stopped, the screen went back to the menu, and the system shut it self down as if I had told it to.

My idle temps with the case open in a 70 F room are
35 C for CPU
45 C for MB
and the CPU fan is spinning (stock fan/heatsink) at 1300rpm idle, 1550-1600 when under the stress tests

according to PC Probe and Everest.
when I run the system stability test with Everest selecting on stress CPU & FPU, the temp for the CPU hovers around 50-51 C peaking at 52-53, and the mboard gets up to 51-52 C.

I haven't updated the bios yet, as i've never done that before so i'm not quite sure how to. But I doubt that'll fix this heating issue.

 

[Maximilian]

Your heat spreader may not be flat, it may need filing down at the sides. Many core2 owners have found that the heat spreader is concave and have filed down the sides to make the entire surface flat. Your CPU may be one of these concave ones, theres a whole thread about it, ill go see if i can dig it up.

Read this thread

and check out how its done

Go put a straight edge to the cpu and make sure its flat first though obviously.

 

[spectre8]

wow. that is a pretty extreme measure to take on something I just spent $250 on lol.

from the looks of it though, the original poster on that thread had much higher idle temps. He had a 55 C idle, thats nearly 20 C more than my cpu right now.

If anything i'm more concerned with the fact that my Mboard is hovering at 49-50C with just windows running, and an open case.


I also don't think that is the problem with my system because the CPU's temperature drops rather rapidly once I stop the everest stress test, dropping within seconds from 52 C down to 41, and slowly from there. The original poster on that other thread had 70+ C as peak temps for when the system was on load. I'm no where near those readings.

 

[phile]

I'm running an E6600 on a P5W DH with Zalman CNPS9500 LED, and after weeks of deliberation I have finally decided I'm going to lap both the CPU and HSF. My current temps are satisfactory w/ a 25% o/c, but I'm just too damn curious to see how much the lapping will improve my temps. We shall soon see. I could not find a single hardware store in my area that ells high-grit sandpapers, so I ordered a lapping kit from PCVIPER. Once it's all done, I will post my before and after temps. The results I get from the lapping should be a good case for, or against, lapping when not dealing with excessively high temps. For the most part, I'm doing this because my inner-geek really wants to see a flat, mirror-finished, copper heatspreader.

Stay tuned...

-phil

 

[spectre8]

I just don't think that would fix my problem though.

The system can clearly run stable until I decide to use the graphics card basically.
My question is WHY!?!? lol.

 

[phile]

Originally posted by: spectre8
I just don't think that would fix my problem though.

The system can clearly run stable until I decide to use the graphics card basically.
My question is WHY!?!? lol.

Your problems do not seem to be related to overheating. Are you using onboard audio, or a dedicated soundcard? If you're using a dedicated soundcard, have you disabled onboard audio in the BIOS?

Whether you're using onboard audio, or a dedicated soundcard, try running the same tests with sound disabled. You can disable your audio in the device manager.

-phil

 

[spectre8]

i'm using onboard audio. the only card in my system right now is the video card.


last night I did lower the... frequency? I think it is, of the CPU from 266 to 200, which brought the overall speed down to 1.6ghz rather then 2.16ghz. I was then able to play GTA : San Andreas for as long as I liked, but the computer continued to beep (the system speaker) every 15-20 seconds.

 

[phile]

Originally posted by: spectre8
i'm using onboard audio. the only card in my system right now is the video card.


last night I did lower the... frequency? I think it is, of the CPU from 266 to 200, which brought the overall speed down to 1.6ghz rather then 2.16ghz. I was then able to play GTA : San Andreas for as long as I liked, but the computer continued to beep (the system speaker) every 15-20 seconds.

Try running at stock (266) with the onboard audio disabled in the BIOS. TRoubleshooting is a process of systematically ruling out each one of your components. Onboard audio has a history of causing problems, so it's as good a place as any to start.

-phil

 

[cmdrdredd]

Originally posted by: phile
I'm running an E6600 on a P5W DH with Zalman CNPS9500 LED, and after weeks of deliberation I have finally decided I'm going to lap both the CPU and HSF. My current temps are satisfactory w/ a 25% o/c, but I'm just too damn curious to see how much the lapping will improve my temps. We shall soon see. I could not find a single hardware store in my area that ells high-grit sandpapers, so I ordered a lapping kit from PCVIPER. Once it's all done, I will post my before and after temps. The results I get from the lapping should be a good case for, or against, lapping when not dealing with excessively high temps. For the most part, I'm doing this because my inner-geek really wants to see a flat, mirror-finished, copper heatspreader.

Stay tuned...

-phil

You could have found the paper necessary from the autoparts store. it's nade by 3M and called Imperial wetordry automotive sandpaper.

 

[Geekwannab]

Ok. I got my Ninja and I have mounted it with AS5. I have currently OCed it a tad(280FSB) , Vcore 1.3V with the RAM doing 1:1 and am still running it without the case. SP2004 is running @ home as I post this.The last time I checked it it was like 6 hours and counting. Ambient is still 20C. Here are my temps

IDLE

PC Probe - 26C
Everest and Speedfan-27C
Intel TAT and coretemp -39-40C

LOAD

PC Probe - 35-37C
Intel TAT and coretemp -46-48C

I am happy with the temps now without any lapping. woot

 

[Geekwannab]

Screenies
Idle
Load

 

[Geekwannab]

going for 300 FSB with 1.3 Vcore as 280 FSB is stable with good temps

 

[phile]

Originally posted by: Geekwannab
going for 300 FSB with 1.3 Vcore as 280 FSB is stable with good temps

With my E6600/P5WDH I was able to go as high as FSB 320 with vcore at auto. I'm currently at FSB 333 with vcore at 1.35.

-phil

 

[Geekwannab]

I was SP2004 stable @333FSB,Vcore@auto. But the problem was the load temps was like 52C in TAT, I dropped the FSB to default (266) and vcore to auto and load temps was like 49C,which means the bios is giving more than the default voltages. I am now on a mission to find out how my E6600 responds to votages

 

[phile]

Originally posted by: Geekwannab
I was SP2004 stable @333FSB,Vcore@auto. But the problem was the load temps was like 52C in TAT, I dropped the FSB to default (266) and vcore to auto and load temps was like 49C,which means the bios is giving more than the default voltages. I am now on a mission to find out how my E6600 responds to votages

A load temp of 52C in TAT is fanatstic! At 1.35 vcore my cores idle at about 48C. BTW, I've noticed that Orthos stressing both cores at 100% doesn't push temps nearly as high as when stressing the cores at 100% with TAT. At load stressing with Orthos, my core temps peak at about 64C, while stressing with TAT gets them into the low 70s.

-phil

 

[Geekwannab]

Originally posted by: phile

Originally posted by: Geekwannab
I was SP2004 stable @333FSB,Vcore@auto. But the problem was the load temps was like 52C in TAT, I dropped the FSB to default (266) and vcore to auto and load temps was like 49C,which means the bios is giving more than the default voltages. I am now on a mission to find out how my E6600 responds to votages

A load temp of 52C in TAT is fanatstic! At 1.35 vcore my cores idle at about 48C. BTW, I've noticed that Orthos stressing both cores at 100% doesn't push temps nearly as high as when stressing the cores at 100% with TAT. At load stressing with Orthos, my core temps peak at about 64C, while stressing with TAT gets them into the low 70s.

-phil

Thanks for the info. I will try stress tesing with TAT also.I am currently running it without the case, maybe thats whay the temps are that low(38-39C in TAT with default everything is really good), or maybe I got lucky and got a good chip. I am trying to find out how far I can push my E6600 without the case. My goal around 60C load in TAT/core temp. Once I reach that temp limit, nomatter what the OC is I am gonna stop. I am not into benching and I would like to put everything inside a case like any other average person.

 

[phile]

Well, if you're stable at 333 with vcore auto, you have a better chip than mine. I'm curious to know which week yours was manufactured. I have a week 24 chip. IF you don't know how to determine this, look on the retail box sticker where it says FPO/BATCH #: L6******. The two numbers after L6 represent the week of manufacture.

-phil

 

[Geekwannab]

Originally posted by: phile
Well, if you're stable at 333 with vcore auto, you have a better chip than mine. I'm curious to know which week yours was manufactured. I have a week 24 chip. IF you don't know how to determine this, look on the retail box sticker where it says FPO/BATCH #: L6******. The two numbers after L6 represent the week of manufacture.

-phil

week 27